scCODA

scCODA employs a Bayesian model to detect compositional changes in cell type proportions from single-cell RNA sequencing (scRNA-seq) data.

Key Features:

• Bayesian Framework: Implements a Bayesian statistical model that accounts for the compositional nature of single-cell data and mitigates issues arising from low sample sizes.
• Integration with Scanpy: Accepts results from Scanpy-based single-cell analysis workflows to incorporate scCODA's compositional analysis into existing pipelines.
• Enhanced Detection Performance: Improves detection of cell type compositional changes that can be missed by traditional methods.

Scientific Applications:

• Disease and perturbation studies: Detects cell type composition changes associated with diseases and experimental stimuli.
• Developmental biology: Characterizes compositional shifts in cell types during development.
• Cellular dynamics and validation: Identifies experimentally verified cell type changes to inform studies of complex cellular dynamics across conditions.

Methodology:

Uses a Bayesian statistical framework to model compositional single-cell data and infer changes in cell type proportions across conditions, addressing compositionality and low sample-size challenges.